When Micromobility Met Its Limits
Remember when electric scooters were going to fix urban transit forever? Around 2018 and 2019, every major European city was suddenly ankle-deep in rentable e-scooters, and Voi was one of the companies leading that charge across the continent. It was scrappy, fast-moving, and very much a hardware-and-logistics play. Now the founders behind Voi have pivoted about as hard as you can pivot — from physical wheels on city streets to AI agents running inside enterprise software stacks. Their new startup, Pit, just raised $16 million in seed funding, with a16z leading the round.
As someone who spends most of my time building bots and thinking about how automated systems actually behave in production, I find this transition genuinely interesting — not because of the founder pedigree, but because of what Pit is actually trying to build.
What Pit Is Actually Building
Pit is going after enterprise AI with a specific angle: products that learn from clients how their businesses run, then create custom software to automate those workflows. That framing matters. A lot of enterprise AI tooling right now is generic — you get a capable model, some connectors, and a prompt template, and you’re told to figure out the rest. Pit’s stated approach is to start from the client’s own operational knowledge and build outward from there.
From a bot-builder’s perspective, that’s the right instinct. The hardest part of automating any workflow isn’t the AI layer — it’s the context layer. What does this company actually do? What are the edge cases? Where do humans currently make judgment calls that no one has ever written down? Generic tools fail at exactly that point. A system that learns the business first, then builds the automation, has a real shot at getting past the demo stage and into actual daily use.
Why Stockholm Keeps Producing This
Pit is based in Stockholm, which at this point should surprise no one. The Swedish capital has quietly become one of the most productive AI startup cities in Europe. There’s a culture there of building things that work at scale — Spotify, Klarna, King — and that engineering-first mindset seems to be carrying over into the current AI wave. Pit fits that pattern: less hype-driven positioning, more focus on a specific enterprise problem with a concrete product thesis.
The a16z backing is worth paying attention to here too. The firm has been selective about which European AI bets it leads, so Pit clearing that bar suggests the team has articulated something more than a vague “AI for enterprise” pitch. Whether the product lives up to that in practice is a separate question, but the signal is real.
What This Means for Bot Builders
If you’re building bots and automation systems — which is most of what we cover here at ai7bot.com — Pit’s approach is worth studying for a few reasons:
- Context ingestion is the moat. If Pit can build solid pipelines for capturing how a business actually operates, that institutional knowledge becomes the defensible part of the product. The AI model underneath is increasingly a commodity. The context layer is not.
- Custom software generation is the output. Pit isn’t just surfacing insights or building dashboards — it’s generating software. That puts it in a different category from most enterprise AI tools and raises the stakes considerably. Generated code that runs in production needs to be auditable, maintainable, and safe to update.
- Enterprise trust is slow to earn. The scooter business was fast and consumer-facing. Enterprise software sales cycles are long, procurement is cautious, and IT security teams ask hard questions. The founders are moving into a very different rhythm than what built Voi.
The Rage-Bait Footnote
Reports mention that Pit gained some early attention through provocative social media posts. I’ll be honest — that’s a minor red flag for me. Enterprise buyers are not won over by LinkedIn controversy, and a reputation for rage-bait can make procurement conversations harder before they even start. It’s a common early-stage attention tactic, but it’s one the team will probably want to leave behind as they move deeper into serious enterprise deals.
A Startup Worth Watching
Pit is early, the product is still taking shape, and $16 million in seed funding is a starting point, not a finish line. But the core idea — teach the system how your business works, then let it build the automation — is a genuinely useful framing for where enterprise AI needs to go. As someone building in this space every day, I’ll be watching how their architecture evolves and whether the custom software generation holds up under real production conditions. That’s where the interesting technical story will be.
🕒 Published: